Skip to main content

Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

Yes, I think what I did is the most I could do without Nintendo shutting down the project lol. As a "technical demo," I think it can still be protected as a non-profit artistic project... I think if I had done it in Godot 4, the project would have looked much better, since I could have used Vulkan or Direct3D, but back then I did it with Godot 3 which uses OpenGL, and OpenGL isn't exactly the best library for making 3D games, so optimization was a real challenge. Surprisingly, it uses raytracing, which is completely unnecessary, but I wanted to see how far I could push it: It doesn't have baked lighting, it's real-time lighting. I wanted to see how many enemies I could put on screen without frame drops, how many background elements I could add without performance tanking...

Obviously, the CRT effect helps a lot, since it lets me lower the game's resolution to give it that "retro" look, which is nice—I like it (I use it a lot)—but aside from personal taste, it was also an optimization technique. Because if I left the game in 16:9 and 1080p (or higher), not only would it have had frame drops on modest hardware, it wouldn't have looked as good without adding more decoration; it would have looked really empty.

In Godot 4, things change a lot; performance is much more stable with Vulkan or Direct3D. Even if I had added a lot more decoration and enemies, it would still run smoother.

But that's how it is—part of the magic isn't always adding more stuff, but making it look good with less. Still, it's necessary to do these kinds of experiments to learn the limits of the tools. 👍

Unless you tested it—at length—behind the scenes, I wouldn’t think anywhere on-screen in World 1-1 would come close to tripping up a modern game engine. Even the Nintendo barely lost its stuff in, 8-1, where you had 2 bullet bill launchers, 2 Goombas (or Busy Beetles) and a Koopa Paratroopa hopping around. Unless you meant something else entirely, and I didn’t catch on lol.

I didn't mean that Super Mario was a complex challenge lol, any simple game would have done, the simpler the better. What I wanted to test was the visual/performance aspect. Besides, Super Mario is the quintessential platformer—if you can make a 3D version of Super Mario and make it look good, then you can make any 3D platformer. That's usually how I test engines; I don't complicate things with prototypes, I do what I already know well so I can focus on what I'm actually interested in testing 😝

When I have time, I'd like to take a look at Defold, since apparently you can already do interesting things in 3D with it, and I'll probably end up doing something similar.

Oh okay. I understand now. Defold? I just looked it up. Sounds like it could be an interesting alternative to Godot. Until I am able to get my computer upgraded, am on the lookout for decent alternatives to Unreal/Unity. Well, good luck with your next experiment.

I haven't tried it, but I do want to give it a shot. I've followed its evolution; it's an engine that has gradually been evolving in 3D, which is ultimately what interests me the most. Obviously, you can't expect Unreal Engine-level results, but then again, that's not what I'm looking for either. My dream game would have a stylized design, and right now Defold seems to be achieving some interesting results in 3D games. I think it would be the perfect time to take a look, because it's an engine known for being very well optimized, and it's always good to know about alternatives.